The documentation 
at http://doc.akka.io/docs/akka/snapshot/scala/cluster-usage.html doesn't 
state that the PoisonPill performs differently than the behaviour listed 
here 
<http://doc.akka.io/docs/akka/snapshot/scala/routing.html#note-router-terminated-children-scala>,
 
so I assumed if I used a Broadcast message like below:

...
  def vectorizeUserContent(streamer : ActorRef, libraryId : Long, 
senderNode : ActorRef) = {

    val userVectorizationReducers =
      context.actorOf(FromConfig.props(Props[ UserVectorizationReducer ]), 
"userVectorizationReducer")

    context.actorOf(Props(new Actor {

      override def preStart() {
        context.watch(userVectorizationReducers)
        streamer ! StartStream(libraryId)
      }

      override def receive = {
        case dto : RawUserContentDto => userVectorizationReducers ! 
ConsistentHashableEnvelope(dto, dto.userId)
        case EndOfStream() => {
          userVectorizationReducers ! 
Broadcast(ReturnUserVectors(senderNode))
          *userVectorizationReducers ! Broadcast(PoisonPill)*
        }
        case Terminated(userVectorizationReducers) => {
          senderNode ! VectorizationCompleted()
          self ! PoisonPill
        }
      }

    }), "userContentVectorizerDirector")
  }
...

where userVectorizationReducers is a remotely deployed pool of routees, the 
Terminated message would be retrieved due to the watch made in preStart(). 
Unfortunately, the broadcasted PoisonPill does not trigger the expected 
Terminated message I'm waiting for. If I however replace the broadcast in 
my 'EndOfStream()' case with userVectorizationReducers ! PoisonPill, then 
the Terminated message is returned. In both cases, all routees are 
terminated individually (evident through some logging in the postStop() 
method of the UserVectorizationReducer actor). I'm assuming the PoisonPill 
is actually being routed to the routees where they can be gracefully 
shutdown after all other messages have been processed, but I could be wrong.

Can I get clarification if this is in fact true and it is the recommended 
way to shutdown a pool of cluster routees?

Cheers,
Sean

-- 
>>>>>>>>>>      Read the docs: http://akka.io/docs/
>>>>>>>>>>      Check the FAQ: 
>>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html
>>>>>>>>>>      Search the archives: https://groups.google.com/group/akka-user
--- 
You received this message because you are subscribed to the Google Groups "Akka 
User List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to akka-user+unsubscr...@googlegroups.com.
To post to this group, send email to akka-user@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.

Reply via email to